Are all TILs annotated in the ROIs ? ¶
By: 1810929 on May 2, 2022, 2:07 p.m.
We aren't quite sure if there are bounding boxes for all the TILs in the Datasets or if just some examples are given. Any tipps would be appreciated.
By: vaibhav.s on May 10, 2022, 9:40 a.m.
Can we get confirmation regarding this? 1. Its seems partial annotation for tils within the ROIs provided while visualizing the provided annotated data. 2. Also there is unlabelled class in the ROI segmentation annotated data among 7 classes.
By: mart.vanrijthoven on May 12, 2022, 7:16 a.m.
Dear Vaibhav Singh,
1. There should be no partial annotations within ROIs. Could you please give an example of partial annotations?
2. There are indeed unlabeled pixels in the ROI. For example, some pixels between two adjacent tissue types are not labeled. Non-labeled pixels can also exist in TCGA slides because an 'exclude' label was present in the original annotations. Furthermore, in the TCGA dataset, some ROIs are rotated. Therefore, the extracted rectangle region also includes pixels that are not labeled. The unlabeled pixels should not be interpreted as another class but should be treated as non-labeled pixels.
Best wishes, Mart